R Notebook
Bundestagswahl 2017
Soziodemografische Analyse auf Wahlkreisebene
Katharina Brunner und Christian Endt, Süddeutsche Zeitung
Quelle: Bundeswahlleiter/Statistisches Bundesamt/Bundesagentur für Arbeit Stand: Montag 25.09.2017, vielzufrüh
To-Do
- Hartz IV Ausländer: Prozent oder je 1000?
- Achsen und Viertellinien beschriften
- Weitere Hypothesen eintragen (auch: Wo ist es vielleicht interessant, dass es KEINE Korrelation gibt?)
- Export implementieren (SVG, WWW, MOB)
- Party-Colors mit jenen in der SZ-Config zusammenlegen
- Warum gibt es NAs in den sozidemografischen Zahlen?
Übersicht: Korrelationen 2013
correlations %>% count(strength)## # A tibble: 5 x 2
## strength n
## <chr> <int>
## 1 medium/substantial 2
## 2 strong 1
## 3 very weak 243
## 4 weak to moderate 12
## 5 <NA> 6significant_correlations <-
correlations %>%
filter(strength != "very weak") %>%
arrange(desc(abs(corr))) %>%
tbl_df
select(significant_correlations,-strength)## # A tibble: 15 x 3
## partei indikator corr
## <fctr> <chr> <dbl>
## 1 fdp wahlbeteiligung 0.6267437
## 2 grüne wahlbeteiligung 0.5284472
## 3 linke wahlbeteiligung -0.5066514
## 4 spd Reli_Katholisch.Prozent 0.2638097
## 5 spd HartzIV__Ausländer 0.2557419
## 6 spd Betreute_Kinder.je1000Einw -0.2535921
## 7 spd Ohne_Migrationshintergrund.Prozent -0.2439009
## 8 spd Mit_Migrationshintergrund.Prozent 0.2439009
## 9 linke Betreute_Kinder.je1000Einw 0.2323338
## 10 union Svp_Beschäftigte_LandwForstwFischerei.Prozent 0.2240593
## 11 spd Svp_Beschäftigte_LandwForstwFischerei.Prozent -0.2204676
## 12 spd Haushaltseinkommen.jeEinw 0.2165344
## 13 spd Ausländeranteil.Prozent 0.2159620
## 14 linke Handwerksunternehmen.je1000Einw 0.2080676
## 15 linke Alter_Ü75.Prozent 0.2048453Print Scatterplots
some selected ideas
scatter("spd","Reli_Katholisch.Prozent","SPD und Anteil der Katholiken")## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
scatter("spd","Mit_Migrationshintergrund.Prozent","SPD und Einwohner mit Migrationshintergrund")## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
scatter("afd","Abitur.Prozent","AfD und Abiturientenquote")## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
scatter("afd","Ausländeranteil.Prozent","AfD und Ausländeranteil")## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
scatter("fdp","Haushaltseinkommen.jeEinw","FDP und Einkommen")## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
scatter("fdp","wahlbeteiligung","FDP und Wahlbeteiligung")
scatter("spd","wahlbeteiligung")
Significant correlations (cor >= 0.2)
for (i in 1:nrow(significant_correlations)){
print(
scatter(
as.character(significant_correlations[[i,1]]),
as.character(significant_correlations[[i,2]])
)
)
}
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 24 rows containing non-finite values (stat_smooth).## Warning: Removed 24 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 24 rows containing non-finite values (stat_smooth).## Warning: Removed 24 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
## Warning: Removed 2 rows containing non-finite values (stat_smooth).## Warning: Removed 2 rows containing missing values (geom_point).## Warning: Removed 598 rows containing missing values (geom_vline).
HANDLE WITH CARE: All possible combinations
this will make your machine groan…
# for (i in unique(dfdf$indikator)){
# indikatorr <- i
# for (j in unique(dfdf$partei)){
# parteyy <- j
# print(
# scatter(
# as.character(parteyy),
# as.character(indikatorr)
#
# )
# )
# }
# }